Recently added
Plugins grouped by the day they were added to the directory, newest first.
Sep 18, 2026
- ketch638stars
Optional Claude Code integration for the ketch CLI: wires up the `ketch mcp serve` MCP server (search, code, docs, scrape, crawl, tag) and installs the bundled ketch research skill. The CLI stays the zero-infrastructure path — this plugin is a convenience, never a prerequisite. Requires the ketch binary >= v0.10.0 on PATH: brew install ketch or go install github.com/1broseidon/ketch@latest.
Claude Code1 MCP server - vibeshield1stars
Bug-hunting security audit for AI-generated code. One skill, parallel bug-hunter subagents, findings in the VibeShield contract format.
Claude Code1 skill - model-router0stars
Keep the expensive lead model on planning and review; route routine work to pinned cheaper workers (Haiku, Sonnet, Opus) and audit which models actually ran.
Claude Code2 skills - artor0stars
Publish, version, and share prototypes via the artor CLI — init/publish/open/share, fork or restore, and sync org skills, env vars, mock datasets, and registries.
Claude Code1 skill - govkit0stars
GovKit skills for governed, AI-assisted delivery: draft Rules and Gherkin scenarios directly from an evidence-backed opportunity with no epic, user story or estimate required; materialize behavior an approved commitment baseline already selected; define a workflow's L1/L2/L3 views with references into canonical behavior; refine Draft 0 specs (Gherkin + NFRs + evaluation criteria); size and slice scenarios into MVP/V1/V2 releases; validate repo readiness before coding; map a spec corpus into a scored feature map; emit delivery metrics; and generate synthetic test data. Story mapping from a tracker epic remains available as a legacy route.
Claude Code7 skills - aipos-p20stars
AIPOS Pillar 2 (Rapid Validation) skills: test the riskiest assumptions before committing engineering capacity — interview guides, problem sizing, visual prototype briefs, demand tests, feasibility spikes, GenAI eval stubs, and the viability brief that carries the go / no-go / revise Validation Decision.
Claude Code1 skill - aipos-p10stars
AIPOS Pillar 1 (Continuous Discovery) skills: run evidence-governed planning over a Product Definition Graph — read the ranked opportunity list over MCP, grade the evidence behind each candidate, find the coverage and attention gaps, and record the Exploration Decisions (explore with budget and horizon, decline, or defer) that hand off to Pillar 2.
Claude Code1 skill - paraqualis-gxp2stars
Life-sciences regulatory & computer-system-validation toolkit for Claude Code — GAMP 5, 21 CFR Part 11, EU GMP Annex 11, the live eCFR, openFDA, and an IQ/OQ/PQ qualification engine (/qualify:build, /qualify:requirements, /qualify:review). Commands, skills, sub-agents, and a document-protection hook. Run /openfda:setup once to provision a free personal openFDA API key (one minute, raises rate limits ~1,000 → ~120,000 requests/day).
Claude Code2 skills1 MCP server - workflow-lab0stars
Local/interactive workflow skills: setup, iterate, transcript-parser, workflow-tuning. Never required by the cloud pipeline.
Claude Code4 skills - workflow0stars
Cloud-safe core workflow skills for Claude Code: plan, refine, execute, comprehensive-review, close-out, post-build
Claude Code5 skills - vogel-tools0stars
Vogel internal tools for managers. Includes vhp-query for pulling project and ticket data out of the VHP database.
Claude Code2 skills - rn-compliance-analyst0stars
Scan and auto-fix Android/iOS app store policy violations in React Native and Capacitor projects
Claude Code4 skills - wt-secret-guard1stars
Private function-hook secret scrubber for Bash results and pasted prompts
Claude Code - workflow-toolbox1stars
Agents-only shim of the workflow-toolbox plugin, loaded by the observe server into delegated (server-launched) SDK sessions so workflow-toolbox:* agentTypes resolve there despite settingSources: []. Byte-identity-mirrored from plugin/agents/ (launch-agents-identity gate). It ships no context-injecting surfaces and only matcher-narrowed process hooks referenced via ../bin from the parent plugin (no bin/ dir of its own): the verifier-CLI guard on PostToolUse/Bash + PreToolUse/StructuredOutput, and the Path-B envelope-intercept hook on PreToolUse/Agent for workflow-toolbox:opencode-verifier. EVERY OTHER agent (leaf/lean/etc.) runs effectively bare: the hooks spawn only on their matched tools, then self-scope immediately to the narrow bridge case they own. Not for interactive install — the full workflow-toolbox plugin already provides these agents + hooks.
Claude Code - workflow-toolbox1stars
Toolbox for Claude Code's Workflow tool — author deterministic multi-agent workflow scripts (workflow-composer, with the @workflow-toolbox compile-time pattern toolkit as its standard library), scaffold a build-clean workflow skeleton from a spec (toolkit-scaffold), diagnose a failed run from its journal and decide whether resumeFromRunId helps (workflow-debugger), re-verify the runtime surface after a Claude Code upgrade (upgrade-canary), run a bias-free multi-lens analysis of any subject when relevant (independent-analysis), and run a deep-grounding pass that collects and verifies evidence against the real sources before asserting or analyzing (deep-grounding), and install, check, and migrate editable, versioned copies of the plugin's own rules and agent definitions into a project's or user's config directory (adopt). Ships the independent-analysis workflow that skill triggers (domain-agnostic adversarial multi-lens analysis), a Stop hook that auto-surfaces a finished Workflow run's cost + traceability audit report (always a notice; grabs the session when the run looks like trouble, was degraded by a silently-denied tool call, or requested external delegation whose agents show no real external-CLI invocation), and a tracker-agnostic actionability Stop gate that blocks only while startable work remains with nothing running, paired with a Planka producer that writes the snapshot it reads automatically from board-reading tool calls (never a guess: it stays silent on a partial read or on a project with no Depends-on convention). Also ships the pilot agent suite (pilot / pilot-orchestrator / pilot-watchdog) and the pilot-wave composer skill for driving tracked task-tracker cards by delegation, plus a conditional SessionStart hook that injects a generic delegation ladder where a project does tracked/delegated work and points at the delegation-chain skill (a non-user-invocable reference on spawn shapes, the upward report chain, wake signals, and what each shipped watcher/guard does and does not cover) before a delegation decision is made. Ships two Planka kanban skills for any project that wants board-based task tracking: planka-tracking (onboard a project — create the board, migrate an existing progress.md into cards, write the pointer) and what-next (propose the logical next step from that board, or from progress.md/conversation when there's no board), both requiring only a reachable planka MCP server.
Claude Code19 skills1 MCP server Turn a founder's expertise into LinkedIn authority posts tuned by the 360Brew algorithm, with a living client profile and self-enriching memory.
Agent Plugins4 skillsSystem that turns a founder's expertise into authority on LinkedIn. Posts tuned by the 360Brew algorithm, with a living client profile (authority-context.md) and self-enrichment via local memory. For Claude Code and Claude Cowork.
Claude Code18 skills- template-plugin1stars
A minimal working example plugin. Copy this directory to start a new plugin in this marketplace.
Claude Code1 skill - minh-toolkit1stars
Minh's personal skill collection: coaching-style skills for learning, writing, branding, and code, plus a YouTube video critic. Each skill's own SKILL.md documents its purpose in detail.
Claude Code12 skills - alternate-cloud0stars
Deploy and manage workloads on Alternate Clouds (Akash + Spheron + Phala TEE) from any AI coding assistant. Bundles the `ac` CLI reference plus task-specific skills for static sites, Docker apps, templates, raw servers, and troubleshooting.
Claude Code7 skills - iskron1stars
NKS methodology skills — orient, write, design, and weave an NKS graph through the iskron_* MCP tools.
CodexClaude Code21 skills1 MCP server - galaxy-skills0stars
Curated Galaxy skills for running an analysis: connecting to a Galaxy server, the MCP tool surface, dataset collections, user-defined tools, workflow reports and turning an ad-hoc history into a reusable workflow. Mirror of galaxyproject/galaxy-skills (skills/).
CodexClaude Code6 skills - galaxy-mcp0stars
Galaxy MCP server (galaxy-mcp) plus the galaxy-connect setup skill.
CodexClaude Code1 skill1 MCP server - galaxy-dev-skills0stars
Galaxy development skills: tool XML wrappers, Nextflow-to-Galaxy conversion, ToolShed revisions in usegalaxy-tools, UCSC track hubs and Galaxy Hub news posts. Mirror of galaxyproject/galaxy-skills (dev-skills/).
CodexClaude Code5 skills - foundry-skills0stars
Galaxy Workflow Foundry cast skills: schema-validated, corpus-grounded steps for turning papers, Nextflow pipelines and CWL workflows into validated Galaxy workflows. Mirror of galaxyproject/foundry (casts/claude/skills).
Codex50 skills - three-role-model1stars
The 3-role development model as a Claude Code plugin: planner -> plan-review -> executor -> execution-review, each a separate subagent, never self-review. Also ships Autonomous Pipeline Mode — a Stop hook that catches stopping to ask permission for an already-approved step, plus a post-compact resume sequencer (ELI5 + 3-tier re-sequence + one approve-gate) and its doctrine. Bundles the role-enforcing hooks, the orchestration skills, a memory-search shim, the standalone doctrine, and a scaffolding command that makes the model the default for new skills, agents, hooks, and commands.
Claude Code5 skills - sdd32stars
Spec-Driven Development: ADRs, specifications, sprint planning, parallel implementation, code review, and documentation generation.
Claude Code21 skills LAL Mastermind AI Paralegal Skills — case intake, financial-discovery intake, core drafting/QC, case roadmap and closing, plus the Core Foundation layer. Florida family law.
Claude Code24 skills- abdallah-skills0stars
Frontend and backend conventions for Next.js + NestJS + Prisma, plus a correction-capture loop.
Claude Code7 skills - lifecycle-axis-sdlc0stars
Enforced AI-native SDLC for Claude Code: the intent to spec to plan to diff to review to incident artifact chain as skills, read-only reviewer subagents, and governance templates and scripts. Deterministic hooks stay repo-local -- install them with scripts/adopt.sh --with-hooks.
Claude Code - catalyst-probes3stars
Consultation and prompt engine for Claude Mods. The module is the core; [probe.<id>] tables in probes.toml are the consultants and [prompt.<id>] tables author a system-prompt section, a tool description or a command description. The only carrier is the mod: a CLAUDE_*_CARRIER handle set to mod or left empty/unset arms the probe here; any other value loudly refuses the calls this probe acts on (the patch-side copy is gone).
Claude Code - catalyst3stars
Catalyst: disciplined agentic development - SDD pipeline with decision-boundary model tiering, spec hardening, premortem gates, verification, and handoffs.
CodexClaude Code14 skills - trail-split0stars
Plan shared gear, food, weight and cost across a group trip so nothing is carried twice and nothing is forgotten.
Claude Code1 skill1 MCP server - daube-agent-bridge0stars
Write one portable AI skill and compile it for multiple agent ecosystems.
Claude Code1 MCP server 20 software design principles and a god-class extraction playbook, as a Claude Code skill.
Claude Code- pi-agent-for-claude31.9kstars
pi agent for claude: a `pi` subagent type whose model steps run the pi CLI, so pi subagents and agent-team teammates live in Claude Code's own agent UI, streamed live, on any model pi can reach.
Claude Code - civil-agreement0stars
Write or revise a project's CLAUDE.md: the working agreement, the levers that differ per team, and the template.
Claude Code1 skill - oci-founder-toolkit0stars
A safe-by-default, backend-first path from an existing codebase to a secure and cost-aware OCI journey.
CodexClaude CodeAgent Plugins1 skill - lessons-learned0stars
Automatic mistake capture and proactive lesson injection for AI coding agents. Mines conversation logs for mistake patterns, structures them as indexed lessons, and injects relevant warnings before the agent repeats the mistake.
Claude Code1 skill1 MCP server - motion-trace0stars
Capture and review web animation as structured, per-frame data (not video). Extracts resolved motion state from the live page so an agent can review easing, timing, overshoot, prefers-reduced-motion accessibility, and real-runtime jank during browsing or E2E. Deterministic capture for CSS/WAAPI plus a realtime mode for JS-library/scroll motion; also a regression diff between two captures.
Claude Code1 skill1 MCP server - flowlever0stars
Local review cockpit for specs and PR review/respond: a fingerprinted findings ledger that tracks issues across rounds, a guided review-flow stepper, exportable work orders, and UI-triggered jobs. Bundles the zero-dependency Node app plus /flowlever:* skills (audit, propose, apply-spec, rework, brief, track, start, stop, watch, poll, pr-review, pr-respond, summarize).
Claude Code13 skills - dor-dod0stars
Specification & development toolkit built on the FZAG DXP Definition of Ready/Done. Three skills sharing one source-of-truth checklist file: spec (author a story/bug/feature spec to the relevant DoR as tool-agnostic markdown), check (audit any artifact against the matching DoR/DoD and report gaps), and promote (publish a markdown draft to an Azure DevOps work item or a Confluence spec page). Works without MCP (markdown in/out) and uses Atlassian/Azure DevOps MCP when present.
Claude Code3 skills - design-eng0stars
Design-engineering skills for reverse-engineering UIs and giving them motion. ui-reverse-engineer turns a screenshot or live URL into a faithful, disassembled reconstruction with an extracted design-token system; motion-system designs the temporal layer (motion tokens, choreography, per-state transitions) for a design, inferring it from taste or measuring it from a video/URL; figma-export pushes the extracted tokens and components into Figma. Together they cover static design extraction, motion, and design-tool round-tripping.
Claude Code2 skills - cmux-swarm0stars
Orchestrate a team of Claude Code agents in cmux: spawn workers, dispatch tasks via brief files, monitor by commit / footer-marker / anchored DONE, and isolate parallel edits with git worktrees. Bundles filtered code-check and visual-check skills for low-token worker reporting, plus a cmux install-if-missing bootstrap command.
Claude Code3 skills - seal-install0stars
Installs Sealmetrics on a site from scratch: creates the site, plans the snippet and the funnel with you, writes them, simulates them against the real tracker before anything ships, then confirms the first hit and each event on the live site. Carries the local MCP server, the only one that exposes the provisioning, planning and verification tools.
Claude Code1 skill1 MCP server - seal-copilot0stars
AI marketing optimization analyst for Sealmetrics consentless analytics. Diagnoses performance, finds revenue opportunities, monitors carts intraday, audits catalog friction per SKU, and surfaces operational waste — all from 100% of your traffic data, no consent sampling.
Claude Code15 skills1 MCP server - seal-copilot0stars
AI marketing optimization analyst for Sealmetrics consentless analytics. Diagnoses performance, finds revenue opportunities, monitors carts intraday, audits catalog friction per SKU, and surfaces operational waste — all from 100% of your traffic data, no consent sampling.
Codex15 skills1 MCP server - sealmetrics-codex0stars
SealMetrics analytics MCP tools and implementation workflow for Codex.
Codex1 skill1 MCP server - transformations59stars
Transform raw dlt pipeline data into a Canonical Data Model. Build an ontology, design a CDM with Kimball dimensional modeling, write @dlt.hub.transformation functions, and validate the output.
Claude Code6 skills1 MCP server - sql-database-pipeline59stars
Build SQL database pipelines with dlt: connect to any SQL source, load tables to a destination, tune performance with backends
Claude Code8 skills1 MCP server